SlideShare a Scribd company logo
1 of 39
Vision – based RunwayVision – based Runway
Recognition foR uaVRecognition foR uaV
autonomous Landingautonomous Landing
Neeraj TiwariNeeraj Tiwari
INDIAN INSTITUTE OF SPACE SCIENCEINDIAN INSTITUTE OF SPACE SCIENCE
AND TECHNOLOGYAND TECHNOLOGY
15May201715May2017
objectiVe and goaLobjectiVe and goaL
►Low cost for UAV.Low cost for UAV.
►Light weight.Light weight.
►Unmanned System.Unmanned System.
►Easily Installation.Easily Installation.
LandingLanding
Human operator
for high level
control
Laptop Computer for
Vision Processing
and Control
Algorithms
Ground Station
bLock diagRambLock diagRam
RC Plane with Camera
30 frames per second
720x480 pixels RGB
Downsample to 360x240
Remote Control
(72 Mhz)
Analog Video
(NTSC 900 Mhz)
oVeRViewoVeRView
►System for landing a UAV on a runwaySystem for landing a UAV on a runway
 Small RC airplaneSmall RC airplane
 Only sensor is a fixed, forward-looking cameraOnly sensor is a fixed, forward-looking camera
 Finds the runway using Hough Transform.Finds the runway using Hough Transform.
 Linear control systemLinear control system
►ExperimentsExperiments
 Microsoft Flight Simulator (no flight model)Microsoft Flight Simulator (no flight model)
 Partial implementation on a real UAVPartial implementation on a real UAV
aiRPoRt tRackingaiRPoRt tRacking
gPs tRackinggPs tRacking
HoRiZon detectionHoRiZon detection
►The horizon profile shape can be used forThe horizon profile shape can be used for
attitude determination and other localizationattitude determination and other localization
process.process.
►This method is based on edge basedThis method is based on edge based
horizon detection or image segmentationhorizon detection or image segmentation
based horizon detection methodbased horizon detection method
►This technique work when vehicle is closeThis technique work when vehicle is close
to the Ground and there is a strong horizonto the Ground and there is a strong horizon
edge.edge.
HoRiZon detectionHoRiZon detection
►An edge strength histogram is computedAn edge strength histogram is computed
based on the edge strength produced by thebased on the edge strength produced by the
canny edge detector and the top p% of thecanny edge detector and the top p% of the
points are obtained as possible points thatpoints are obtained as possible points that
comprise the horizon.comprise the horizon.
►The Standard Hough Transform is thenThe Standard Hough Transform is then
applied to fit probable lines in the binaryapplied to fit probable lines in the binary
edge map.edge map.
►The horizon will be the largest edgeThe horizon will be the largest edge
strength.strength.
HoRiZon detection matLabHoRiZon detection matLab
outPutoutPut
HoRiZon detection matLabHoRiZon detection matLab
outPutoutPut
HORIZON DETECTION WITH HIGHHORIZON DETECTION WITH HIGH
PROCESSING SYSTEMPROCESSING SYSTEM
MaIN STEPSMaIN STEPS
1.1. Locate the runway in each video frameLocate the runway in each video frame
2.2. Estimate the attitude of the UAVEstimate the attitude of the UAV
3.3. Steer the UAV towards the runwaySteer the UAV towards the runway
maintaining the correct glideslopemaintaining the correct glideslope
LOCaTE THE RuNWaYLOCaTE THE RuNWaY
IMAGE
DILATION
IMAGE
THRESHOL
DING
EDGE
DETECTIO
N
SMALL
REGION
RMOVAL
CONVOLUTIO
N
OPERATION
HOUGH
TRANSFOR
M
IDENTIFY PEAK
AND EXTRACT
LINES
RUNWAY
IMAGE
SUPERIMPOS
E LINES ON
THE
ORIGINAL
RUNWAY
IMAGE
IMaGE DILaTIONIMaGE DILaTION
►Dilation adds pixels to the boundaries of theDilation adds pixels to the boundaries of the
objects in an image.objects in an image.
►The number of pixels added or removedThe number of pixels added or removed
from the objects in an image depends onfrom the objects in an image depends on
the size and shape of the structuringthe size and shape of the structuring
element used in image process.element used in image process.
►The runway markings are highlighted usingThe runway markings are highlighted using
this technique.this technique.
IMaGE DILaTIONIMaGE DILaTION
ORIGINAL IMAGE DILATED IMAGE
IMaGE THRESHOLDINGIMaGE THRESHOLDING
►Thresholding is the process of mappingThresholding is the process of mapping
pixels to produce a two level image.pixels to produce a two level image.
►Here OTSU’s thresholding appears moreHere OTSU’s thresholding appears more
appropriate.appropriate.
►OTSU’s method chooses the thresholdOTSU’s method chooses the threshold
value in such a way that the intra classvalue in such a way that the intra class
variance of black and white pixels isvariance of black and white pixels is
minimized.minimized.
IMaGE THRESHOLDINGIMaGE THRESHOLDING
GRAY IMAGE OTSU’s THRESHOLDING
THRESHOLDINGTHRESHOLDING
GRAY IMAGE LOCAL THRESHOLDING
EDGE DETECTIONEDGE DETECTION
►Sobel edge operator is selected for thisSobel edge operator is selected for this
operation because of its optimal solution foroperation because of its optimal solution for
low defective edge rate , localization oflow defective edge rate , localization of
edge and giving one response for singleedge and giving one response for single
edge.edge.
EDGE DETECTIONEDGE DETECTION
CANNY OPERATION SOBEL OPERATOR
REMOVAL OF SMALL REGIONREMOVAL OF SMALL REGION
MAGE BEFORE SMALL REGION REMOVAL
AFTER SMALL REGION REMOVAL
LINE DETECTION TECHNIQUESLINE DETECTION TECHNIQUES
►It used to extract the pair of linesIt used to extract the pair of lines
representing the runway boundaries.representing the runway boundaries.
►Once the runway boundaries are roughlyOnce the runway boundaries are roughly
obtained, straight line detection is to beobtained, straight line detection is to be
performed.performed.
►Image convolution can be used to easilyImage convolution can be used to easily
detect lines, only vertical, 45’ and 135’detect lines, only vertical, 45’ and 135’
masks are sufficient.masks are sufficient.
LINE DETECTION TECHNIQUESLINE DETECTION TECHNIQUES
-1 2 -1
-1 2 -1
-1 2 -1
-1 -1 2
-1 2 -1
2 -1 -1
2 -1 -1
-1 2 -1
-1 -1 2
VERTICAL LINES
45’ LINES
135’ LINES
CONVOLUTION MASKS FOR STRAIGHT LINE DETECTION
LINE FITTING USING HOUGHLINE FITTING USING HOUGH
TRANSFORMTRANSFORM
►Hough transform is a robust method used toHough transform is a robust method used to
extract arbitrary shapes such as lines ,extract arbitrary shapes such as lines ,
circles, ellipses, out of an image.circles, ellipses, out of an image.
►Straight lines are parameterized in the formStraight lines are parameterized in the form
ρ = x.cos= x.cos ɵ + y.sin+ y.sin ɵ
► In Hough space , lines are mapped to aIn Hough space , lines are mapped to a
point such that a point represents allpoint such that a point represents all
possible lines.possible lines.
LINE FITTING USING HOUGHLINE FITTING USING HOUGH
TRANSFORMTRANSFORM
STACk OF REFERENCE FRAMESSTACk OF REFERENCE FRAMES
ESTIMATE THE UAV ATTITUDEESTIMATE THE UAV ATTITUDE
► 6 Degrees of Freedom6 Degrees of Freedom
 Pitch, Bank, Heading, Elevation, Distance, CoursePitch, Bank, Heading, Elevation, Distance, Course
► StrategyStrategy
 IgnoreIgnore DistanceDistance
 FindFind PitchPitch andand BankBank from the horizon line (x-axis)from the horizon line (x-axis)
 FindFind ElevationElevation,, HeadingHeading,, CourseCourse from the runwayfrom the runway
INTUITIVE GEOMETRyINTUITIVE GEOMETRy
►Relationship between runway appearanceRelationship between runway appearance
and UAV attitudeand UAV attitude
 This is how human pilots land visuallyThis is how human pilots land visually
Too High On Target Too Far Right
FORMAL GEOMETRyFORMAL GEOMETRy
► 3D Projection3D Projection
 C = Internal CalibrationC = Internal Calibration
 R = External CalibrationR = External Calibration
► Small Angle ApproximationSmall Angle Approximation
 Assume the UAV is flying smooth and levelAssume the UAV is flying smooth and level
ESTIMATE THE UAV ATTITUDEESTIMATE THE UAV ATTITUDE
►Recover the orientation parametersRecover the orientation parameters
►Vanishing point of the runwayVanishing point of the runway
►Beginning of the runwayBeginning of the runway
Control the UAVControl the UAV
► Cascaded Linear FeedbackCascaded Linear Feedback
ControllerController
 Two separate chainsTwo separate chains
 Two gainsTwo gains
► ProportionalProportional
► IntegralIntegral
► IntuitiveIntuitive
 If UAV is too far right, steer leftIf UAV is too far right, steer left
 If UAV is too high, pitch downIf UAV is too high, pitch down
 Bank angle is derivative ofBank angle is derivative of
heading, heading is derivativeheading, heading is derivative
of courseof course
 Pitch is derivative of elevationPitch is derivative of elevation
PI 1
PI 1
PI 2
PI 3
PI 2
Course
Heading
Bank
Elevation
Pitch
AUtopilot GUiAUtopilot GUi
AlGorithm performAnCeAlGorithm performAnCe
►Multiple stagesMultiple stages
 Control loops run at 50 HzControl loops run at 50 Hz
►Integrates smoothly even while input stays sameIntegrates smoothly even while input stays same
 Horizon detection runs at 10 HzHorizon detection runs at 10 Hz
►Pitch and bank are the most sensitivePitch and bank are the most sensitive
 Runway detection runs at 2 HzRunway detection runs at 2 Hz
►Elevation and course are the least sensitiveElevation and course are the least sensitive
SimUlAtor reSUltSSimUlAtor reSUltS
►Microsoft Flight SimulatorMicrosoft Flight Simulator
►Simulator-in-the-loop (separate computer)Simulator-in-the-loop (separate computer)
 ICRA08_1140_VI_fi.mp4ICRA08_1140_VI_fi.mp4
 Horizon 2007 09 Sep 11 Tue 08.13pm.aviHorizon 2007 09 Sep 11 Tue 08.13pm.avi
SimUlAtor reSUltSSimUlAtor reSUltS
► Error from earth curvatureError from earth curvature
ACtUAl UAV experimentSACtUAl UAV experimentS
►Only using partial implementationOnly using partial implementation
 Horizon stabilizationHorizon stabilization
 Road following (no runway available)Road following (no runway available)
 Only brief periods of autonomous controlOnly brief periods of autonomous control
reAl time VieWreAl time VieW
ConClUSionSConClUSionS
►Successful but imprecise landingsSuccessful but imprecise landings
►Performance is applicable to ARDUINOPerformance is applicable to ARDUINO
 Slower and more stable than actual UAVsSlower and more stable than actual UAVs
►Assumption of linear system is notAssumption of linear system is not
applicable near the runwayapplicable near the runway
 This is why the aircraft oscillates before landingThis is why the aircraft oscillates before landing
►Future workFuture work
 Incorporate flight model into controller designIncorporate flight model into controller design
thAnK-YoUthAnK-YoU

More Related Content

Similar to Autonomous UAV Landing

Hidden lines & surfaces
Hidden lines & surfacesHidden lines & surfaces
Hidden lines & surfacesAnkur Kumar
 
Precission Approach Path Indicator (PAPI)
Precission Approach Path Indicator (PAPI)Precission Approach Path Indicator (PAPI)
Precission Approach Path Indicator (PAPI)Suhanto Tuban
 
Unreal Fest 2023 - Lumen with Immortalis
Unreal Fest 2023 - Lumen with ImmortalisUnreal Fest 2023 - Lumen with Immortalis
Unreal Fest 2023 - Lumen with ImmortalisOwen Wu
 
Visual odometry & slam utilizing indoor structured environments
Visual odometry & slam utilizing indoor structured environmentsVisual odometry & slam utilizing indoor structured environments
Visual odometry & slam utilizing indoor structured environmentsNAVER Engineering
 
Kenya Nasu: IVUS controlled re-entry in antegrade CTO PCI
Kenya Nasu: IVUS controlled re-entry in antegrade CTO PCIKenya Nasu: IVUS controlled re-entry in antegrade CTO PCI
Kenya Nasu: IVUS controlled re-entry in antegrade CTO PCIEuro CTO Club
 
3D stage for upright microscope
3D stage for upright microscope3D stage for upright microscope
3D stage for upright microscopeDr. Pranav Rathi
 
Artificial Neural Network Based Object Recognizing Robot
Artificial Neural Network Based Object Recognizing RobotArtificial Neural Network Based Object Recognizing Robot
Artificial Neural Network Based Object Recognizing RobotJaison Sabu
 
Optimisation of drawframe autoleveller
Optimisation of drawframe autolevellerOptimisation of drawframe autoleveller
Optimisation of drawframe autolevellerHitesh Choudhary
 
Lane detection by use of canny edge
Lane detection by use of canny edgeLane detection by use of canny edge
Lane detection by use of canny edgebanz23
 
The International Journal of Engineering and Science (IJES)
The International Journal of Engineering and Science (IJES)The International Journal of Engineering and Science (IJES)
The International Journal of Engineering and Science (IJES)theijes
 
The International Journal of Engineering and Science
The International Journal of Engineering and ScienceThe International Journal of Engineering and Science
The International Journal of Engineering and Sciencetheijes
 
Dr.Kawewong Ph.D Thesis
Dr.Kawewong Ph.D ThesisDr.Kawewong Ph.D Thesis
Dr.Kawewong Ph.D ThesisSOINN Inc.
 
Deep Learning A-Z™: AutoEncoders - Module 6
Deep Learning A-Z™: AutoEncoders  - Module 6Deep Learning A-Z™: AutoEncoders  - Module 6
Deep Learning A-Z™: AutoEncoders - Module 6Kirill Eremenko
 
Types of scanners
Types of scannersTypes of scanners
Types of scannersPramoda Raj
 

Similar to Autonomous UAV Landing (20)

eGWM
eGWMeGWM
eGWM
 
Arte Y Robotica
Arte Y RoboticaArte Y Robotica
Arte Y Robotica
 
Hidden lines & surfaces
Hidden lines & surfacesHidden lines & surfaces
Hidden lines & surfaces
 
Precission Approach Path Indicator (PAPI)
Precission Approach Path Indicator (PAPI)Precission Approach Path Indicator (PAPI)
Precission Approach Path Indicator (PAPI)
 
Unmanned Ground Vehicle
Unmanned Ground VehicleUnmanned Ground Vehicle
Unmanned Ground Vehicle
 
Unreal Fest 2023 - Lumen with Immortalis
Unreal Fest 2023 - Lumen with ImmortalisUnreal Fest 2023 - Lumen with Immortalis
Unreal Fest 2023 - Lumen with Immortalis
 
Visual odometry & slam utilizing indoor structured environments
Visual odometry & slam utilizing indoor structured environmentsVisual odometry & slam utilizing indoor structured environments
Visual odometry & slam utilizing indoor structured environments
 
Kenya Nasu: IVUS controlled re-entry in antegrade CTO PCI
Kenya Nasu: IVUS controlled re-entry in antegrade CTO PCIKenya Nasu: IVUS controlled re-entry in antegrade CTO PCI
Kenya Nasu: IVUS controlled re-entry in antegrade CTO PCI
 
UNIT-V
UNIT-VUNIT-V
UNIT-V
 
3D stage for upright microscope
3D stage for upright microscope3D stage for upright microscope
3D stage for upright microscope
 
Artificial Neural Network Based Object Recognizing Robot
Artificial Neural Network Based Object Recognizing RobotArtificial Neural Network Based Object Recognizing Robot
Artificial Neural Network Based Object Recognizing Robot
 
Optimisation of drawframe autoleveller
Optimisation of drawframe autolevellerOptimisation of drawframe autoleveller
Optimisation of drawframe autoleveller
 
Nadia2013 research
Nadia2013 researchNadia2013 research
Nadia2013 research
 
Lane detection by use of canny edge
Lane detection by use of canny edgeLane detection by use of canny edge
Lane detection by use of canny edge
 
The International Journal of Engineering and Science (IJES)
The International Journal of Engineering and Science (IJES)The International Journal of Engineering and Science (IJES)
The International Journal of Engineering and Science (IJES)
 
The International Journal of Engineering and Science
The International Journal of Engineering and ScienceThe International Journal of Engineering and Science
The International Journal of Engineering and Science
 
Dr.Kawewong Ph.D Thesis
Dr.Kawewong Ph.D ThesisDr.Kawewong Ph.D Thesis
Dr.Kawewong Ph.D Thesis
 
report
reportreport
report
 
Deep Learning A-Z™: AutoEncoders - Module 6
Deep Learning A-Z™: AutoEncoders  - Module 6Deep Learning A-Z™: AutoEncoders  - Module 6
Deep Learning A-Z™: AutoEncoders - Module 6
 
Types of scanners
Types of scannersTypes of scanners
Types of scanners
 

Recently uploaded

Online banking management system project.pdf
Online banking management system project.pdfOnline banking management system project.pdf
Online banking management system project.pdfKamal Acharya
 
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...Call Girls in Nagpur High Profile
 
data_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfdata_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfJiananWang21
 
Work-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxWork-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxJuliansyahHarahap1
 
notes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.pptnotes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.pptMsecMca
 
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...Arindam Chakraborty, Ph.D., P.E. (CA, TX)
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlysanyuktamishra911
 
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night StandCall Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Standamitlee9823
 
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Bookingroncy bisnoi
 
Bhosari ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For ...
Bhosari ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For ...Bhosari ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For ...
Bhosari ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For ...tanu pandey
 
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...roncy bisnoi
 
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...Call Girls in Nagpur High Profile
 
UNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its PerformanceUNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its Performancesivaprakash250
 
Block diagram reduction techniques in control systems.ppt
Block diagram reduction techniques in control systems.pptBlock diagram reduction techniques in control systems.ppt
Block diagram reduction techniques in control systems.pptNANDHAKUMARA10
 
Design For Accessibility: Getting it right from the start
Design For Accessibility: Getting it right from the startDesign For Accessibility: Getting it right from the start
Design For Accessibility: Getting it right from the startQuintin Balsdon
 
Unleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leapUnleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leapRishantSharmaFr
 
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Bookingdharasingh5698
 

Recently uploaded (20)

Online banking management system project.pdf
Online banking management system project.pdfOnline banking management system project.pdf
Online banking management system project.pdf
 
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
 
data_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfdata_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdf
 
Work-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptxWork-Permit-Receiver-in-Saudi-Aramco.pptx
Work-Permit-Receiver-in-Saudi-Aramco.pptx
 
notes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.pptnotes on Evolution Of Analytic Scalability.ppt
notes on Evolution Of Analytic Scalability.ppt
 
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
Navigating Complexity: The Role of Trusted Partners and VIAS3D in Dassault Sy...
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghly
 
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night StandCall Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
Call Girls In Bangalore ☎ 7737669865 🥵 Book Your One night Stand
 
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Wakad Call Me 7737669865 Budget Friendly No Advance Booking
 
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Ramesh Nagar Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
 
Bhosari ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For ...
Bhosari ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For ...Bhosari ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For ...
Bhosari ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For ...
 
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
 
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
 
UNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its PerformanceUNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its Performance
 
Block diagram reduction techniques in control systems.ppt
Block diagram reduction techniques in control systems.pptBlock diagram reduction techniques in control systems.ppt
Block diagram reduction techniques in control systems.ppt
 
Design For Accessibility: Getting it right from the start
Design For Accessibility: Getting it right from the startDesign For Accessibility: Getting it right from the start
Design For Accessibility: Getting it right from the start
 
Unleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leapUnleashing the Power of the SORA AI lastest leap
Unleashing the Power of the SORA AI lastest leap
 
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 BookingVIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
VIP Call Girls Ankleshwar 7001035870 Whatsapp Number, 24/07 Booking
 
Water Industry Process Automation & Control Monthly - April 2024
Water Industry Process Automation & Control Monthly - April 2024Water Industry Process Automation & Control Monthly - April 2024
Water Industry Process Automation & Control Monthly - April 2024
 
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
 

Autonomous UAV Landing

  • 1. Vision – based RunwayVision – based Runway Recognition foR uaVRecognition foR uaV autonomous Landingautonomous Landing Neeraj TiwariNeeraj Tiwari INDIAN INSTITUTE OF SPACE SCIENCEINDIAN INSTITUTE OF SPACE SCIENCE AND TECHNOLOGYAND TECHNOLOGY 15May201715May2017
  • 2. objectiVe and goaLobjectiVe and goaL ►Low cost for UAV.Low cost for UAV. ►Light weight.Light weight. ►Unmanned System.Unmanned System. ►Easily Installation.Easily Installation.
  • 4. Human operator for high level control Laptop Computer for Vision Processing and Control Algorithms Ground Station bLock diagRambLock diagRam RC Plane with Camera 30 frames per second 720x480 pixels RGB Downsample to 360x240 Remote Control (72 Mhz) Analog Video (NTSC 900 Mhz)
  • 5. oVeRViewoVeRView ►System for landing a UAV on a runwaySystem for landing a UAV on a runway  Small RC airplaneSmall RC airplane  Only sensor is a fixed, forward-looking cameraOnly sensor is a fixed, forward-looking camera  Finds the runway using Hough Transform.Finds the runway using Hough Transform.  Linear control systemLinear control system ►ExperimentsExperiments  Microsoft Flight Simulator (no flight model)Microsoft Flight Simulator (no flight model)  Partial implementation on a real UAVPartial implementation on a real UAV
  • 7. HoRiZon detectionHoRiZon detection ►The horizon profile shape can be used forThe horizon profile shape can be used for attitude determination and other localizationattitude determination and other localization process.process. ►This method is based on edge basedThis method is based on edge based horizon detection or image segmentationhorizon detection or image segmentation based horizon detection methodbased horizon detection method ►This technique work when vehicle is closeThis technique work when vehicle is close to the Ground and there is a strong horizonto the Ground and there is a strong horizon edge.edge.
  • 8. HoRiZon detectionHoRiZon detection ►An edge strength histogram is computedAn edge strength histogram is computed based on the edge strength produced by thebased on the edge strength produced by the canny edge detector and the top p% of thecanny edge detector and the top p% of the points are obtained as possible points thatpoints are obtained as possible points that comprise the horizon.comprise the horizon. ►The Standard Hough Transform is thenThe Standard Hough Transform is then applied to fit probable lines in the binaryapplied to fit probable lines in the binary edge map.edge map. ►The horizon will be the largest edgeThe horizon will be the largest edge strength.strength.
  • 9. HoRiZon detection matLabHoRiZon detection matLab outPutoutPut
  • 10. HoRiZon detection matLabHoRiZon detection matLab outPutoutPut
  • 11. HORIZON DETECTION WITH HIGHHORIZON DETECTION WITH HIGH PROCESSING SYSTEMPROCESSING SYSTEM
  • 12. MaIN STEPSMaIN STEPS 1.1. Locate the runway in each video frameLocate the runway in each video frame 2.2. Estimate the attitude of the UAVEstimate the attitude of the UAV 3.3. Steer the UAV towards the runwaySteer the UAV towards the runway maintaining the correct glideslopemaintaining the correct glideslope
  • 13. LOCaTE THE RuNWaYLOCaTE THE RuNWaY IMAGE DILATION IMAGE THRESHOL DING EDGE DETECTIO N SMALL REGION RMOVAL CONVOLUTIO N OPERATION HOUGH TRANSFOR M IDENTIFY PEAK AND EXTRACT LINES RUNWAY IMAGE SUPERIMPOS E LINES ON THE ORIGINAL RUNWAY IMAGE
  • 14. IMaGE DILaTIONIMaGE DILaTION ►Dilation adds pixels to the boundaries of theDilation adds pixels to the boundaries of the objects in an image.objects in an image. ►The number of pixels added or removedThe number of pixels added or removed from the objects in an image depends onfrom the objects in an image depends on the size and shape of the structuringthe size and shape of the structuring element used in image process.element used in image process. ►The runway markings are highlighted usingThe runway markings are highlighted using this technique.this technique.
  • 16. IMaGE THRESHOLDINGIMaGE THRESHOLDING ►Thresholding is the process of mappingThresholding is the process of mapping pixels to produce a two level image.pixels to produce a two level image. ►Here OTSU’s thresholding appears moreHere OTSU’s thresholding appears more appropriate.appropriate. ►OTSU’s method chooses the thresholdOTSU’s method chooses the threshold value in such a way that the intra classvalue in such a way that the intra class variance of black and white pixels isvariance of black and white pixels is minimized.minimized.
  • 17. IMaGE THRESHOLDINGIMaGE THRESHOLDING GRAY IMAGE OTSU’s THRESHOLDING
  • 19. EDGE DETECTIONEDGE DETECTION ►Sobel edge operator is selected for thisSobel edge operator is selected for this operation because of its optimal solution foroperation because of its optimal solution for low defective edge rate , localization oflow defective edge rate , localization of edge and giving one response for singleedge and giving one response for single edge.edge.
  • 20. EDGE DETECTIONEDGE DETECTION CANNY OPERATION SOBEL OPERATOR
  • 21. REMOVAL OF SMALL REGIONREMOVAL OF SMALL REGION MAGE BEFORE SMALL REGION REMOVAL AFTER SMALL REGION REMOVAL
  • 22. LINE DETECTION TECHNIQUESLINE DETECTION TECHNIQUES ►It used to extract the pair of linesIt used to extract the pair of lines representing the runway boundaries.representing the runway boundaries. ►Once the runway boundaries are roughlyOnce the runway boundaries are roughly obtained, straight line detection is to beobtained, straight line detection is to be performed.performed. ►Image convolution can be used to easilyImage convolution can be used to easily detect lines, only vertical, 45’ and 135’detect lines, only vertical, 45’ and 135’ masks are sufficient.masks are sufficient.
  • 23. LINE DETECTION TECHNIQUESLINE DETECTION TECHNIQUES -1 2 -1 -1 2 -1 -1 2 -1 -1 -1 2 -1 2 -1 2 -1 -1 2 -1 -1 -1 2 -1 -1 -1 2 VERTICAL LINES 45’ LINES 135’ LINES CONVOLUTION MASKS FOR STRAIGHT LINE DETECTION
  • 24. LINE FITTING USING HOUGHLINE FITTING USING HOUGH TRANSFORMTRANSFORM ►Hough transform is a robust method used toHough transform is a robust method used to extract arbitrary shapes such as lines ,extract arbitrary shapes such as lines , circles, ellipses, out of an image.circles, ellipses, out of an image. ►Straight lines are parameterized in the formStraight lines are parameterized in the form ρ = x.cos= x.cos ɵ + y.sin+ y.sin ɵ ► In Hough space , lines are mapped to aIn Hough space , lines are mapped to a point such that a point represents allpoint such that a point represents all possible lines.possible lines.
  • 25. LINE FITTING USING HOUGHLINE FITTING USING HOUGH TRANSFORMTRANSFORM
  • 26. STACk OF REFERENCE FRAMESSTACk OF REFERENCE FRAMES
  • 27. ESTIMATE THE UAV ATTITUDEESTIMATE THE UAV ATTITUDE ► 6 Degrees of Freedom6 Degrees of Freedom  Pitch, Bank, Heading, Elevation, Distance, CoursePitch, Bank, Heading, Elevation, Distance, Course ► StrategyStrategy  IgnoreIgnore DistanceDistance  FindFind PitchPitch andand BankBank from the horizon line (x-axis)from the horizon line (x-axis)  FindFind ElevationElevation,, HeadingHeading,, CourseCourse from the runwayfrom the runway
  • 28. INTUITIVE GEOMETRyINTUITIVE GEOMETRy ►Relationship between runway appearanceRelationship between runway appearance and UAV attitudeand UAV attitude  This is how human pilots land visuallyThis is how human pilots land visually Too High On Target Too Far Right
  • 29. FORMAL GEOMETRyFORMAL GEOMETRy ► 3D Projection3D Projection  C = Internal CalibrationC = Internal Calibration  R = External CalibrationR = External Calibration ► Small Angle ApproximationSmall Angle Approximation  Assume the UAV is flying smooth and levelAssume the UAV is flying smooth and level
  • 30. ESTIMATE THE UAV ATTITUDEESTIMATE THE UAV ATTITUDE ►Recover the orientation parametersRecover the orientation parameters ►Vanishing point of the runwayVanishing point of the runway ►Beginning of the runwayBeginning of the runway
  • 31. Control the UAVControl the UAV ► Cascaded Linear FeedbackCascaded Linear Feedback ControllerController  Two separate chainsTwo separate chains  Two gainsTwo gains ► ProportionalProportional ► IntegralIntegral ► IntuitiveIntuitive  If UAV is too far right, steer leftIf UAV is too far right, steer left  If UAV is too high, pitch downIf UAV is too high, pitch down  Bank angle is derivative ofBank angle is derivative of heading, heading is derivativeheading, heading is derivative of courseof course  Pitch is derivative of elevationPitch is derivative of elevation PI 1 PI 1 PI 2 PI 3 PI 2 Course Heading Bank Elevation Pitch
  • 33. AlGorithm performAnCeAlGorithm performAnCe ►Multiple stagesMultiple stages  Control loops run at 50 HzControl loops run at 50 Hz ►Integrates smoothly even while input stays sameIntegrates smoothly even while input stays same  Horizon detection runs at 10 HzHorizon detection runs at 10 Hz ►Pitch and bank are the most sensitivePitch and bank are the most sensitive  Runway detection runs at 2 HzRunway detection runs at 2 Hz ►Elevation and course are the least sensitiveElevation and course are the least sensitive
  • 34. SimUlAtor reSUltSSimUlAtor reSUltS ►Microsoft Flight SimulatorMicrosoft Flight Simulator ►Simulator-in-the-loop (separate computer)Simulator-in-the-loop (separate computer)  ICRA08_1140_VI_fi.mp4ICRA08_1140_VI_fi.mp4  Horizon 2007 09 Sep 11 Tue 08.13pm.aviHorizon 2007 09 Sep 11 Tue 08.13pm.avi
  • 35. SimUlAtor reSUltSSimUlAtor reSUltS ► Error from earth curvatureError from earth curvature
  • 36. ACtUAl UAV experimentSACtUAl UAV experimentS ►Only using partial implementationOnly using partial implementation  Horizon stabilizationHorizon stabilization  Road following (no runway available)Road following (no runway available)  Only brief periods of autonomous controlOnly brief periods of autonomous control
  • 37. reAl time VieWreAl time VieW
  • 38. ConClUSionSConClUSionS ►Successful but imprecise landingsSuccessful but imprecise landings ►Performance is applicable to ARDUINOPerformance is applicable to ARDUINO  Slower and more stable than actual UAVsSlower and more stable than actual UAVs ►Assumption of linear system is notAssumption of linear system is not applicable near the runwayapplicable near the runway  This is why the aircraft oscillates before landingThis is why the aircraft oscillates before landing ►Future workFuture work  Incorporate flight model into controller designIncorporate flight model into controller design